An architecture for automatically developing secure OLAP applications from models

Decision makers query enterprise information stored in Data Warehouses (DW) by using tools (such as On-Line Analytical Processing (OLAP) tools) which use specific views or cubes from the corporate DW or Data Marts, based on the multidimensional modelling. Since the information managed is critical, security constraints have to be correctly established in order to avoid unauthorized accesses. In previous work we have defined a ModelDriven based approach for developing a secure DWs repository by following a relational approach. Nevertheless, is also important to define security constraints in the metadata layer that connects the DWs repository with the OLAP tools, that is, over the same multidimensional structures that final users manage. This paper incorporates a proposal to develop secure OLAP applications into our previous approach: improves a UML profile for conceptual modelling; defines a logical metamodel for OLAP applications; and ∗Corresponding author Email addresses: [email protected] (Carlos Blanco), [email protected] (Ignacio Garćıa-Rodŕıguez de Guzmán), [email protected] (Eduardo Fernández-Medina), [email protected] (Juan Trujillo) Preprint submitted to Information and Software Technology November 4, 2014 defines and implements transformations from conceptual to logical models, and from logical models to the secure implementation into a specific OLAP tool (SQL Server Analysis Services).